By John Rezabek, Contributing Editor
If you've been pondering wireless, you've probably read the surveys, many of which show that while a significant percentage of users are deploying wireless transmitters, a large percentage are not. (See Control's wireless survey from last year at www.controlglobal.com/Media/2011/1108/CG1108_CovStry-Wireless.pdf.).
Some of those holding off say they're "waiting on standards convergence." That could be a long wait. Another segment is "waiting on the right application."
There's also the question of start-up costs. Wireless networks don't come cheap. There's also an excellent chance your host vendor is in the "bad guys" camp relative to your wireless choice, so some less-than-elegant systems configuration work is a probable scope item, such as Modbus mapping or OPC. A site survey alone can potentially add a five-figure line item to your project's budget. Then, when we price the actual transmitters and other accoutrements, the capital hurdle becomes a tough one to clear.
There's also what I'd like to christen "Herman's Postulate," which goes something like this: "When I find a measurement worth making I can usually justify running a wire to it." A corollary would be, "If I can get a wire (cable) to it, it's better than a radio."
Think about your household's Ethernet network. If you recently added a player capable of steaming content from the Internet, are you happy with its 802.11b/g wireless connectivity and performance? Or like me, do you get a younger and more agile family member (hint: not your spouse or significant other) to drag a length of Lowe's Cat5 cable through the nasty crawl space under the floor? Maybe you'd rather plunk down $149 for a "Draft 802.11n"-capable router which has a slim chance of improving your radio bandwidth. That's just laziness. Now, I'm not sneezing at laziness. It's the mother of many useful inventions and innovations. But, laziness is frequently not the optimum path for the enterprise that employs us. Why would we consume "mobile" bandwidth for stuff that's not moving around?
Herman's Postulate says "usually" because there are bonafide cases where a wire can't be run—railcars, for example. There are other moving measurements, such as rotary kiln temperature, that we just couldn't get before. But we can get them now if the sensors can be hooked up to a wireless network. There are also some plants where the combination of crushing engineering overhead with an expensive or unreliable labor resource makes avoiding the whole copper deal worth the money.
What if the capital barriers to employing wireless were dramatically less? What if reasonably reliable sensors could be had for a price comparable to or less than a wired one?
Enter the "lick-n-stick" sensor. These are sensors that are cheap and easy to deploy, requiring minimal effort and labor. They're already out there for non-critical applications, such as corrosion monitoring, steam traps, heat tracing, safety showers and pump vibrations. Emerson's new WirelessHART acoustic transmitters, at about $1000 each, are one example. If the cost is less than the per-transaction limit on my corporate VISA card, I can now have a few on the shelf for the "lick-'n-stick" applications that arise.
Wireless begins to be more compelling for measurements for which we'd rarely buy a permanent instrument. Those start-up costs now potentially can be spread across more applications. It's an easier sell to management if I can promise more measurements for the same money. There are already products being sold for these more non-critical applications. This is where cheapness and ubiquity can win in the "fitness for purpose" debate.
I've seen a few devices you could call a lick-n-stick wireless sensor, so I think our suppliers have heard us. They're off the drawing board, so maybe some are headed to a site near you.